H10 clicks when there are flashes on TV

[crusader_apc]

This was originally posted on APC forums on 10/10/2009

---

Hi,

I have APC H10 and Panasonic 50'' plasma TV connected to it. H10 unit shows that load is ~20%. Everything is fine, but when TV shows a movie with flashes (i.e. screen becomes completely white for a moment) -- H10 emits loud click to compensate for voltage jump.
It is kind of ok -- after all unit is supposed to regulate voltage, but in movies with a lot of flashes it gets really annoying. The only way (I found) to avoid this is to set voltage tolerance level to maximum.
It feels like sudden jump in electricity consumption (to render display completely white) drops voltage (or confuses H10 unit).

Can you advise any other way, please? Should I do anything at the electrical box? (like cleaning contact of related circuit/etc)

Michael.

[Erasmus_apc]

This was originally posted on APC forums on 10/12/2009

---

Depending on the type of display you have, current draw can vary in terms of a rated maximum. I doubt the voltage is what's changing, voltage doesn't change when a device needs more power and the AV series units are pretty strict in terms of their voltage output...more current is pulled. This can cause electrical circuits in a building, or a device, to need to compensate for the sudden increase. This is pretty much the concept behind why a bathroom light will dim slightly when a hair dryer is turned on or when an AC is enabled.

The only things you could do are make sure there are no loose contacts in any of the outlets/connections of the entire setup. A short or loose connection can sometimes cause odd sounds. Also, the settings on the display could affect draw and lessen the problem. A lot of plasma TVs come pre-configured for a "SHOP" or "Demo" mode that draws more power than a "Home/Movie" mode. If it doesn't detract from the viewing experience, adjusting this setting would make a small difference in power draw.

Overall though, plasmas draw more power than LCDs do to their need to charge ALL the gas inside the display to turn it into plasma for light generation, unlike an LCD where individual piecex/pixels are on/off. As long as your load and TV are not being turned off this sounds like a symptom of current fluctuation.

[crusader_apc]

This was originally posted on APC forums on 10/13/2009

---

Thanks for helping me!

TV is already calibrated for darker environment it is in.

Assuming that at any given time electrical network provides load of N Watt, then from simple formula 1 W = 1 A * 1 V, it is clear that if electrical current suddenly jumps -- voltage will fall to compensate. Of course there are all sorts of devices on the network that try to compensate (like throwing more coal into the furnace 🙂 ) that but they do not react immediately.
So, for me it is clear why voltage jumps when TV screen suddenly turns white. And H10 itself indicates a jump.

I am looking for any suggestion on how to improve the situation? Like doing some magic at the box, or reducing resistance of the circuit that brings current to given outlet (by any means -- thicker wire, better contacts, magic in the box, etc) -- anything that will help me to understand phenomena and fix it.

[Erasmus_apc]

This was originally posted on APC forums on 10/13/2009

---

That equation and idea is generally correct for this situation, if you are not accounting for power factor to calculate the wattage, etc. different types of protection devices, or a gradual increase that can be compensated for (as you said, compensation by other pieces of the circuit can happen). However also keep in mind that higher amounts of current require higher voltages for efficiencies sake to reduce the current on electrical wiring/circuitry/panels (which is why companies don't make a 12000 watt, 120v single phase input UPS, why dryers use 220v, and why when you raise the amount of draw/clock speed on a video card or processor in a PC, you often need to increase the DC rail voltage). I should clarify as: the AV unit's output voltage shouldn't change very much at all since it has Automatic Voltage Regulation capability (tap transformer that boosts/trims the output in response to input issues), in combination with high quality circuitry and capacitors to provide ride through and further regulation.

In respect to the input to unit, yes, the increase in current could cause the circuit/unit to sag out and basically simulate the effect of dropped voltage. When this click happens, do you see any LED changes on the front of the H10 unit? One thing you could try is the following procedure, which changes how aggressively the unit compensates for high/low input voltages:

To set the AVR range:
1. Push and hold the SEQ/AVR push button for three seconds;
all LEDs are extinguished, indicating the H10 is in delay
programming mode. The LED corresponding to the current
AVR range setting flashes (Narrow, Normal, or Wide).
2. To change AVR range, press the SEQ/AVR push button
until the LED corresponding to the desired setting flashes.
After releasing the push button, the unit returns to normal
status and the normally illuminated LEDs are lit.

The AVR Range adjustments have the following value:
• Narrow AVR Range: 102-132 Volts,
Regulates to: 120 Volts + 5%
• Normal AVR Range: 97-139 Volts,
Regulates to: 120 Volts + 10%
• Wide AVR Range: 92-145 Volts,
Regulates to: 120 Volts + 15%

Also, do you know what your typical input voltage from the wall outlet is? If it's low already, then a sudden jump in current sagging the circuit can make the issue even worse. APC can't recommend any modifications or work to electrical panels, outlets, etc., since that is the realm of licensed electricians.

[crusader_apc]

This was originally posted on APC forums on 10/14/2009

---

When this click happens, do you see any LED changes on the front of the H10 unit?

Yes, as I said before -- H10 indicates voltage drop (on LED panel), and to "work this around" I increased voltage jump tolerance to max. But what is the point of having AVR?
So I am looking for answers in my electrical system.

Also, do you know what your typical input voltage from the wall outlet is?

No idea, but according to LED panel it is 120V. I'll probably need some tester to find out exactly what it is.

APC can't recommend any modifications or work to electrical panels, outlets, etc., since that is the realm of licensed electricians.

Oh, come on, no one will blame APC if you propose a sensible explanation why this may happen. Most of simple homes have similar electrical system and I am sure there is a very simple explanation to this.

[Erasmus_apc]

This was originally posted on APC forums on 10/14/2009

---

The point of AVR is to continually output the nominal voltage of 120v. Otherwise, if input is good, the unit just passes power through at whatever voltage the utility is, and filters it. So, if the voltage drops down enough to engage AVR, the unit will boost by 12%, and then if it goes lower, it will "double boost" the input by 24% to maintain output around 120v. Similar functionality applies to "trimming" high voltage. When this current surge happens, the unit is probably detecting a sag in voltage and the loud click is AVR engaging to maintain output voltage. By "widening" or setting the AVR for its most lenient settings, this would hopefully happen less.

Simple advice I suppose would be lessening total load on the H10, lessening load on the particular circuit that feeds that outlet/area (especially if there's already an AC, fridge, electric fireplace, etc.), and checking for damaged or worn wiring to the outlet/from the circuit panel. Also, if the wiring out from the panel is not large enough, naturally the resistance will be higher, increasing the potential for a voltage drop due to current surge. If you don't have any other issues in the house with dimming lights or faulty equipment, I believe you can safely rule out issues past your house (neighborhood transformer, supply from a power line). I would still bring the issue up to a contractor or licensed electrician, however.

[crusader_apc]

This was originally posted on APC forums on 10/15/2009

---

Indeed, wiring is old and lights dim momentarily whenever something power-hungry turns on. As far as I know this particular circuit does not feed any other outlets (i.e. no fridge, fireplace, etc), but it needs checking.

Also, if the wiring out from the panel is not large enough, naturally the resistance will be higher, increasing the potential for a voltage drop due to current surge.

This is the answer I was looking for... 🙂

I would still bring the issue up to a contractor or licensed electrician, however.

Yes, this sounds sensible...

Thanks!

[crusader_apc · Accepted solution]

This was originally posted on APC forums on 11/28/2009

---

Sorry to reanimate this old thread... I've recently spent some time refreshing my memory on electricity concepts and probably found an explanation to that based on assumption that input voltage (at the service box) is constant regardless of load on the circuits (in my house) -- it should be ok, simply because network's capacity is considerably greater than my meager +/-1kW of load variations.

Attached scheme corresponds to a normal connection representing variable load like TV (Rload) on the same circuit with constant load like light bulbs (Rother) where circuit itself has an impedance (as in my example -- old aluminum wiring with bad/oxidized contacts) marked Rd on diagram. Because Rd is not parallel to Rload -- it produces a 'voltage drop' and as it turned out if appliance draws more power -- voltage drop increases.
Once I have realized that it was relatively easy to produce a calculation (see attached xls) that can estimate Rd and all other aspects of the circuit (with given assumptions). All is derived from simple formulas:
W=I*U
U=I*R
parallel connection:
1/R = 1/R1 + 1/R2
I = I1 + I2
U = U1 = U2
consecutive connection:
R = R1+R2
I = I1 = I2
U = U1 + U2

The idea is to make two measurements on voltage that exists on the load terminals at two predefined wattage levels (assuming it is possible, i.e. appliance reduces resistance until it gets current that generates necessary wattage) -- and then, knowing voltage at the service panel, two voltage measurements at the load, and wattages in both cases it is possible to calculate all other attributes of the circuit.
In my case I made all those assumption and estimated that W1 = 650, W2 = 150, voltage drops 20V and 10V respectively and and there is 120V at the box. This gives me (among other things) ~2Ohms of the unwanted impedance (1 Ohm in each direction). Even if my estimations are wrong -- it is comforting to know exactly what happens -- you know what to look after.

And btw -- I know that aluminum wiring should be touched by licensed electrician only. 🙂